Introduction

There is a growing insight that type 1 diabetes patients and their disease differ [1], even despite receiving the current standard of care to achieve glycemic control. Therefore, a great, and largely unmet, need exists to identity markers and correlates of type 1 diabetes patient subpopulations with differential disease progression, preservation in beta cell function, and glycemic control. This is essential for the design and allocation of personalized and precision medicine strategies, including future immune intervention therapy and the prognosis of disease progression [2].

Gene variation may be used to serve such a purpose, as genetic risk scores have been developed underscoring this notion [3]. Genetic variation in the insulin gene (INS) is the second largest contributor to genetic risk to type 1 diabetes (T1D) [4]. Differences have been identified in a variable number of tandem repeats (VNTR) in the INS promotor region from which the highest risk is associated with class I VNTR alleles, whereas class III VNTRs have been linked to a dominantly protective effect. These INS polymorphisms are believed to influence the expression of proinsulin, where increased thymic expression and decreased pancreatic expression associate with the protective class III VNTR haplotype; this contributes to improved central immune tolerance to proinsulin and protection from type 1 diabetes [5, 6]. This INS promoter polymorphism is in strong linkage disequilibrium with single nucleotide polymorphisms (SNPs) at the 3’ untranslated region (UTR) of the insulin gene [4], allowing to determine the INS-related risk by using tagging SNPs for the promoter polymorphisms and vice versa [7–9].

Yet, INS associated protection is not complete and around 20% of type 1 diabetes patients carries a copy of the ‘protective’ INS variant. Strikingly, pediatric patients developing type 1 diabetes, despite carrying this protective INS variant and receiving the current standard of clinical care, showed reduced insulin autoimmunity, preserved beta cell function, higher C-peptide levels, and improved glycemic control [9]. This points to the possibility that genetic variation may contribute to diagnosis of disease variants and patient stratification.

It has been well established that intensive insulin treatment compared to conventional therapy can prevent and delay diabetes complications [10]. We hypothesized that type 1 diabetes patients receiving this current standard of care and carrying a copy of the protective INS variant have a further reduced risk of developing diabetes complications compared to patients only carrying the susceptible INS variant.

Materials and methods

The association between INS variation and the probability of microvascular diabetes complications, proliferative diabetic retinopathy (PDR) and diabetic kidney disease (DKD) was investigated in participants of the Diabetes Control and Complications Trial and Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study [10]. Following project approval, DCCT/EDIC clinical and SNP study data on 1,441 individuals was downloaded from the National Center for Biotechnology Information database of Genotypes and Phenotypes website, under accession number phs000086.v3.p1. The DCCT/EDIC trial contained two treatment arms (intensive insulin therapy vs conventional therapy), and two cohorts (primary prevention and secondary intervention) within each arm (four groups in total). The association between INS variation and complication probability was analyzed for each group separately. The patients in the primary prevention cohort had T1D for 1–5 years without evidence of retinopathy and albumin secretion rates of <40mg/24h. The patients in the secondary intervention cohort had T1D for 1–15 years and at least one microaneurysm in either eye and albumin secretion rates no greater than 200mg/24h. PDR was defined as any individual with an Early Treatment Diabetic Retinopathy Study score of 12 or more or receiving scatter laser treatment. DKD was defined as any individual with an estimated glomerular filtration rate less than 60 mL/min/1.73m² on two or more consecutive visits, or an albumin excretion rate greater than 300 mg/24h on one visit. The INS tagging SNP was determined by genotyping rs3842752 (also known as +1127PstI) in the 3’ UTR of INS; the A allele represents the protective and G the susceptible variant. SNP data was available for 1,363 individuals. Since our a priori predictions were directional (i.e., protection reduces risk), statistical significance was determined by a one-sided log rank test with a significance level of p<0.05. Further details of the study cohort are provided in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. DCCT/EDIC patient characteristics by insulin gene variant.

https://doi.org/10.1371/journal.pone.0280872.t001

Results

We report that patients carrying protective INS have a further reduced risk of developing diabetes complications compared to those lacking this gene variant. In the primary prevention group of the DCCT/EDIC study, none out of 67 intensively treated patients carrying the protective INS variant (homo- or heterozygous) developed PDR or DKD with follow-up of 20.4 ± 1.6 years after diagnosis. However, 11 out of 258 patients (4.3%) lacking the protective INS variant developed complications (PDR only n = 5, DKD only n = 3, or both PDR and DKD n = 3) in 20.4 ± 1.8 years of follow-up after diagnosis (Fig 1). In the secondary intervention group, intensively treated patients with the protective INS variant showed significantly less PDR (4 of 83 [4.8%]; 26.1 ± 3.9 years post-diagnosis follow up) compared to patients with only the susceptible INS variant (31 of 260 [11.9%]; 26.3 ± 4.1 years post-diagnosis follow up; p = 0.032). However, there were no differences in the DKD frequencies between those with (5 of 83 [6.0%]) and without (11 of 260 [4.2%]) the protective INS variant (Fig 2). None of the seven patients homozygous for the protective INS variant developed DKD 22 ± 1 years after diagnosis. In 695 conventionally treated patients, complication frequencies did not differ between those with or without the protective INS variant (primary prevention group: 3 of 66 developed PDR (5%) vs. 19 of 292 (7%), respectively, and 4 of 66 developed DKD (6%) vs. 14 of 292 (5%), respectively; secondary intervention group: 13 of 59 developed PDR (22%) vs. 72 of 278 (26%), respectively, and 11 of 59 developed DKD (19%) vs. 34 of 278 (12%), respectively. Interestingly, none of the eight patients homozygous for the protective INS variant developed DKD up to 25.2 ± 4.9 years of post-diagnosis follow up.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Association between INS variation and complication probability during intensive insulin treatment.

The complication free probability of PDR (left) and DKD (right) is shown for type 1 diabetes patients carrying the protective (blue) and susceptible (red) INS variant in the primary prevention group of the DCCT/EDIC intensively treated study arm. None of the individuals carrying the protective INS variant developed PDR or DKD (0 of 67 and 0 of 67, respectively); however, PDR and DKD occurred in those carrying the susceptible INS variant (8 of 258 and 6 of 258, respectively). Homozygous and heterozygous protective patients were combined due to low numbers of protective homozygotes.

https://doi.org/10.1371/journal.pone.0280872.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Association between INS variation and complication probability during secondary intervention.

The complication free probability of PDR (left) and DKD (right) is shown for type 1 diabetes patients carrying the protective (blue) and susceptible (red) INS variant in the secondary intervention group of the DCCT/EDIC intensively treated study arm. Significantly fewer individuals carrying the protective INS variant developed PDR (4 of 83) compared to those with the susceptible INS variant only (31 of 260). The probability for developing DKD was not different between individuals carrying the protective INS variant (5 of 83) compared to those with the susceptible INS variant only (11 of 260). Homozygous and heterozygous protective patients were combined due to low numbers of protective homozygotes. Statistical significance was determined using a one-sided log rank test and indicated if p<0.05.

https://doi.org/10.1371/journal.pone.0280872.g002

Discussion

We conclude that the protective INS variant further decreases the low risk for diabetes complications accomplished through intensive insulin treatment and glycemic control in patients with European ancestry. We propose that the combination of immune tolerance to insulin [5, 6] and preserved beta cell function [9] in patients carrying the protective INS variant contributes to superior glycemic control previously reported in carriers with recent onset pediatric T1D, thereby reducing the risk of development of diabetic complications. No complications were reported for patients carrying a protective INS variant in the primary prevention group. The lack of difference in complications frequencies between INS variants in the ‘conventionally’ treated group underscores that intensive treatment, which is the current standard of care, is necessary to reduce complications. The relationship between glycemic control and the association between INS variation and diabetic complication remains unclear. In the DCCT cohort, HbA1c levels were much better in the intensive treatment arm compared to the conventional treatment arm but did not differ between patients with and without the protective INS variant within the intensive treatment arm. Yet, in the Hvidøre cohort an association between superior glycemic control and the protective INS variant was observed in children in the first year after their T1D diagnosis [9]. We speculate that the impact of the protective INS variant on complication probability involves epigenetic modifications early after disease onset (‘metabolic memory’) and may change with improved glycemic control later in disease progression [12]. Future and independent validation of our findings in other large cohorts therefore remains warranted.

We concede that the proportion of type 1 diabetes patients receiving the current standard of care (‘intensive insulin therapy’ in DCCT nomenclature) and nonetheless developing diabetic complications is thankfully relatively small. Yet, we argue that ‘small numbers’ are inherent to diagnosis of type 1 diabetes disease heterogeneity and personalized medicine, and relevant to that minor subset of type 1 diabetes patients involved. Although the DCCT/EDIC cohort is the largest T1D complications cohort with the longest follow-up time, a limitation of this study is the relatively small sample size. Alternative T1D cohorts were considered to replicate our results but these cohorts had even smaller numbers of study subjects, diabetic complications were not (yet) recorded or defined differently or the relevant SNP typing was missing. Fortunately, the Hvidøre pediatric cohort showed proof in the same direction that carrying a copy of the protective INS variant has favorable outcome: in their case on glycemic control and preserved beta cell function [9], both known to significantly reduce or even prevent diabetic complications, and thus indirectly yet independently supporting our observations in the DCCT cohort.

Our new data add to the favorable clinical prognosis in those carrying a copy of the ‘protective’ INS variant thereby linking this ‘endotype’ with superior glycemic control and reduced risk for diabetic complication. This finding may contribute to better selection and stratification of type 1 diabetes patients participating in immune intervention trials, and ultimately personalizing medicine.

Acknowledgments

The authors thank Z. Chen and R. Natarajan for their advice in defining PDR and DKD.